Appropriate Threshold Level of Inflation for Economic ...

Applied Economics and Finance

Vol. 2, No. 3; August 2015

ISSN 2332-7294 E-ISSN 2332-7308

Published by Redfame Publishing

URL: http://aef.redfame.com

127

Appropriate Threshold Level of Inflation for Economic Growth: Evidence

from the Three Founding EAC Countries

Nicas Yabu 1& Nicholaus J. Kessy

1

1Nicas Yabu and Nicholaus J. Kessy, Research Department, Bank of Tanzania, Dar es Salaam, Tanzania.

Correspondence: Nicas Yabu, Bank of Tanzania, 2 Mirambo Street, 11884 Dar es Salaam, Tanzania.

Received: June 12, 2015 Accepted: June 26, 2015 Available online: July 22, 2015

doi:10.11114/aef.v2i3.966 URL: http://dx.doi.org/10.11114/aef.v2i3.966

Abstract

This paper empirically estimated threshold level of inflation, which is conducive for economic growth in the three

founding EAC countries, Kenya, Tanzania and Uganda using panel data set for the period 1970 to 2013. The non-linear

quadratic model was used to estimate the threshold level or the turning point beyond which inflation exerts a negative

impact on economic growth. To examine the inflation-growth relationship other moderating variables were included in

the model. It was found that credit to GDP ratio, degree of openness of the economy and foreign direct investment flows

to EAC member states have significant and positive impact on growth.

In determining threshold level of inflation for the three EAC member states, regression results of the random effect

model establish that the average rate of inflation beyond 8.46 percent has negative and significant impact on economic

growth. For individual countries, findings from the Seemingly Unrelated Regression (SUR), which treats each country

separately, indicate that the optimal levels of inflation for Kenya, Tanzania and Uganda are 6.77 percent, 8.80 percent

and 8.41 percent, respectively, beyond which inflation starts exerting cost on economic growth. The implication for

monetary policy is that policy makers in the EAC member countries need to continue putting effort in achieving and

maintaining single-digit level of inflation to support economic growth.

1. Introduction

1.1 Background

In November 2013, the East African Countries (EAC) signed a protocol for the establishment of a monetary union that

will bring in greater monetary integration, stronger trade, investment and growth. The next move is to establish

independent regional central bank and single currency which is envisaged to bring fourth greater monetary policy

credibility in the region. To ensure that economies going for monetary union are sufficiently integrated with somewhat

similar economic bases, the member states need to fulfill certain agreed upon requirements - the macroeconomic

convergence criteria.

IMF (2012) carried out a study on the convergence criteria and a macroeconomic framework for the EAC in which,

each member state is to achieve a set of goals in a prescribed time period. One of the primary convergence criteria is the

ceiling on headline inflation of 8 percent for which each partner state is to achieve by 2021 and sustain thereafter.

Implicit in the target for inflation is the achievement of high economic growth since the two are fundamental

macroeconomic objectives of most economies. One of the secondary convergence criteria in the EAC region is the

achievement of 7 percent growth rate of output.

In the economic literature, there has been considerable debate on the nature of inflation and growth relationship. There

was an increasing agreement that inflation has distributional effects on long-term economic growth if it gets “too high”

and thus industrial countries consensus was inflation of 2 percent until the 2007 financial crisis that challenged the norm.

Blanchard, DellAricia and Mauro (2010) argue that the effects of inflation on growth are difficult to discern so long as

inflation remains in single digit. They suggest that inflation target of 4 percent for industrial countries might be

appropriate because it leaves some room for expansionary monetary policy in a case of adverse shocks.

For each country or group of countries there exists a certain level or a range of inflation (threshold inflation) which is

conducive for growth. This study attempts to answer a question whether or not the process of achieving the primary

convergence criteria of the ceiling on headline inflation of 8 percent supports economic growth in EAC member states.

Applied Economics and Finance Vol. 2, No. 3; 2015

128

It is from that standpoint this paper intends to determine and examine the threshold level of inflation for the economic

growth among the EAC partner economies.

1.2 Significance of the Study

The relationship between inflation and economic growth play an important role in the EAC. High or unpredictable

inflation rates are regarded as harmful to an overall economy. They add inefficiencies in the market, and make it

difficult for companies to budget or plan long-term. High inflation can cause companies or investors to shift resources

away from high to low inflation countries as a hedge against losses that might be generated from rising costs of inflation.

Estimating the threshold level of inflation will inform the EAC authorities responsible for controlling inflation to ensure

that inflation maintained at levels that spur growth. As growing economies, it is important for the EAC to know the

appropriate threshold level of inflation, beyond which, may be detrimental for sustainable economic growth.

1.3 Scope of the study

This study covered only three EAC founding member states using macroeconomic statistics for a period from 1970 to

2013. These countries have a long historical relationship. They have some common characteristics which can be

generalized for comparison purposes, rather than studying five EAC member countries together, where Burundi and

Rwanda joined the community recently. In the past, Kenya, Tanzania and Uganda enjoyed a long history of co-operation

under different regional integration arrangements. These have included the Customs Union between Kenya and Uganda

in 1917, which the then Tanganyika later joined in 1927; the East African High Commission (1948-1961); the East

African Common Services Organization (1961-1967); the East African Community (1967-1977) and the East African

Co-operation (1993-2000). Also Tanzania, Kenya and Uganda are the three largest economies in the region.

1.4 Organization of the study

Following this introduction, the rest of the paper is structured as follows: Section two discusses the inflation dynamics

and economic growth in the three EAC founding countries. Section three provides both the theoretical and empirical

literature on inflation and economic growth. Section four addresses methodological issue, while section five presents

the analysis and discussion of the empirical results. The last section provides conclusion and policy recommendations.

2. Inflation Dynamics and Economic Growth

The relationship between inflation and growth in the economies of the EAC founding members is evident in the patterns

of historical statistics on inflation and growth.

2.1.1 Inflation Dynamics and Economic Growth in Tanzania

Low inflation has been at the heart of Tanzania‟s monetary authorities since the country attained independence in the

1960s. In fact, the primary mission of the Bank of Tanzania is to maintain domestic price stability that is conducive to

the attainment of macroeconomic stability and the achievement of sustainable growth. The Bank of Tanzania has the

responsibility of ensuring that monetary conditions are consistent with low and suitable inflation (SADC, 2011), (BOT

ACT, 2006).

With the decline of primary commodity prices and the oil crisis of the 1970s, Tanzania experienced a decline in GDP

growth and a sharp increase in the rate of inflation. GDP growth rate declined from an average of 5.7 percent between

1965 and 1975 to an average of 1.8 percent between 1976 and 1979 and then to 1.2 percent between 1980 and 1985;

while at the same time inflation increased from 7.3 to 30.6 percent in the same period (Table 2.1).

Table 2.1 Average Annual Real Rates of Growth (%): Tanzania 1965 - 1985

Period Overall GDP Real Per Capita GDP Inflation Rate

1965 - 1975 5.7 2.5 7.3

1970 - 1976 5.1 1.9 11.1

1976 - 1979 1.8 1.0 14.9

1980 - 1985 1.2 -1.6 30.6

Source: Mbelle, et al, (2002).

The country embarked on economic recovery and structural adjustment programmes from which the economy recorded

a noticeable improvement. Overall, between 1996 and 2009 real GDP growth averaged 6.2 percent, which was higher

than the annual average growth of less than 5 percent in the early 1990s. Over the same period, the rate of inflation was

contained to a single digit, averaging 6.8 percent compared with 25 percent recorded in the early 1990s (BoT, 2011).

For the period between 1995 and 2000 inflation declined steadily while growth largely remained constant. Low and

stable inflation experienced between 2001 and 2005 was reciprocated by a gently rising growth in GDP. In 2012 the


129

economic growth rate was 6.9 percent compared to 6.5 percent recorded in 2010. Meanwhile inflation reached at 16.0

percent in 2012 from 7.2 percent in 2010. Figure 1.1 shows the historical trends of inflation rate and real GDP growth

rate of Tanzania during the period of 1995 to 2013.

Figure 2.1 Trend of inflation and GDP growth rate in Tanzania

Although it is premature to conclude anything on the basis of a visual inspection of figure 2.1, it however more or less

depicts an inverse relationship between rate of inflation and GDP growth rate throughout this period. The relationship

between inflation rate and real growth rate of output for Tanzania is further investigated in Table 2.2 in which the entire

set of actual observations on inflation and growth from 1967 to 2013 is subdivided into ranges starting from the

minimum value of actual rate of inflation in the sample. The second column in the table shows how often (the number

of years) incidences of inflation in a particular range occurred. For example, the inflation range between 2 and 5 percent

occurred eight times between 1980 and 2013. The inflation band between 10 and 20 percent had the highest incidence

of occurrence in the observation period.

Table 2.2 Relationship between Inflation and real GDP growth in Tanzania (1980 – 2013)

Inflation Range / Band Inflation Frequency Mean Inflation Mean GDP Growth

2≤INFL<5 8 4.8 6.7

5≤INFL<10 10 6.8 5.9

10≤INFL<20 13 14.1 3.9

20≤INFL<30 9 25.7 1.8

INFL≥30 9 32.8 3.6

Source: Authors computations

Within each range or band of inflation, the mean inflation and real GDP growth rate is calculated. For example, the

average real GDP growth rate for inflation less than 5 percent is 6.7 percent; 5.9 percent for inflation between 5 – 10

percent etc. The last column in Table 2.2 reveals that mean real GDP growth rate is high at 6.7 in the inflation range less

than 5 percent (mean inflation of 4.8 percent), while the high inflation range is associated with lower GDP growth rates.

This observation signifies that economic activity thrives in an environment of characterized by price stability. That is

low mean inflation rates are associated with high real growth rates of output.

2.1.2 Inflation Dynamics and Economic Growth in Kenya

The first decade following Kenya‟s independence was marked by macroeconomic stability when exchange rate was

fixed and inflation averaged 3 percent. In the 1970‟s, with the first oil price shocks and balance of payments problems,

the rate of inflation began to increase. This increase was accompanied by devaluations and changes in the exchange rate

peg.

One particular turn of events in the 1990‟s was the slowdown in economic growth, the rapid rise in inflation, money

growth and interest rates, and a rapid depreciation of the shilling. Factors behind rapid increase in money supply

0.0

5.0

10.0

15.0

20.0

25.0

30.0

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Per

cen

t

Year

INFL GDP Growth


130

included the foreign aid embargo at the time, escalating fiscal deficits that were financed from money printing and the

exchange rate regime had changed to a dual system in which there was an official exchange rate and a „black market‟

rate.

Over the 15-year period from 1995 to 2010, the growth rate of the economy did not show a steady or regular pattern. In

the year 1995 and 1996, the rate of economic growth in real GDP was 4.3% and 4.0% respectively. However, in the year

1997, the growth rate decreased substantial to a dismal level of 0.2%. The growth rate rose in following years to 4.6%

and 7.0% in 2004 and 2007 respectively (Figure 2.2).

Figure 2.2 Trend of inflation and GDP growth rate in Kenya

Between 1995 and 1996, the government was able to confine underlying inflation within the single digit range. This was

mainly due to tight monetary policy implemented by the monetary authorities in Kenya. However, in 1997, inflation

increased to 11.9%, associated with general election expenditures. Higher inflation is also seen in the year 2000, 2004,

2009 and 2010 mainly being as a result of increasing commodity prices as influenced by rising world oil prices.

From 1995 to 2001, an increase in the real GDP growth does not correspond to decreases in inflation. However,

between the year 2002 and 2004, an increase in real GDP growth rate related to an increase in inflation. At the same

time from 2008 to 2010, the relationship between inflation and economic growth seems to be negative, where a

decrease in the rate of inflation rate from 15.1% in 2008 to 11.6% in 2009 and further to 4.1% was associated with a rise

in the economic growth from 1.5% to 2.6% and further to 5.6% in the same period, respectively. The inflation decreased

further from 10.6% in 2009 to 5.7% in 2013, while growth increased from 2.6% to 4.5% in the same period respectively

(Figure 2.2). Therefore the trend is mixed; hence on the basis of a visual inspection, no general conclusion can be drawn

on the relationship between inflation and economic growth in Kenya.

The historical inflation-growth relationship for Kenya is summarized in Table 2.3 in which average GDP growth is high

when inflation rate is low and high inflation rates are detrimental to growth. There exists an inverse relationship

between inflation and growth in Kenya with a negative 0.5 correlation coefficient.

Table 2.3 Relationship between Inflation and real GDP growth in Kenya (1980 – 2013)

Inflation Range / Band Inflation Frequency Mean Inflation Mean GDP Growth 2≤INFL<5 4 3.4 4.7

5≤INFL<10 11 7.9 4.0 10≤INFL<20 15 12.8 3.6 20≤INFL<30 3 25.6 1.0

INFL≥30 1 46.0 -0.1


2.1.3 Inflation Dynamics and Economic Growth in Uganda

For most of the 1970s and 1980s, Uganda suffered severe macroeconomic imbalances, including high rates of inflation

and balance of payments deficits, because the growth of nominal aggregate demand consistently outstripped the growth

of real supply in the economy. The main reason for this was the printing of money to finance public sector deficits,

leading to large increases in money supply which fuelled high rates of inflation.

In 1987, Uganda embarked on an Economic Recovery Programme with support from the IMF, the World Bank and

other multilateral and bilateral donors. The principal objectives were to revamp the economy and enhance economic

growth, to reduce inflation and to minimize the potential for a balance of payments crisis. Because of the consistency

0.0

5.0

10.0

15.0

20.0

25.0

30.0

1995199619971998199920002001200220032004200520062007200820092010201120122013

Per

cen

t

Year

INFL GDP Growth


131

with which these measures were and are being implemented, real GDP growth rates have been positive since then,

averaging 6.1% per annum from 1986 to 2004. Inflation has been declining overtime from 143.8% to a single digit of

5.0% in the same period. A substantial increase in inflation from 5.0% in 2004 to 14.2% in 2009 was associated with a

significant increase in GDP growth from 6.8% to 7.2% in the same period (Figure 2.3). As in Kenya, the inflation and

economic growth seems to be negative related in Uganda.

Figure 2.3 Trend of inflation and GDP growth rate in Uganda

Table 2.4 summarizes historical data for inflation and GDP growth for Uganda. Again, the tradeoff between inflation

and growth is evident as inflation range from 2 to 10 percent is associated with highest average GDP growth of 7.6

percent.

Table 2.4 Relationship between Inflation and real GDP growth in Uganda (1980 – 2013)

Inflation Range / Band Inflation Frequency Mean Inflation Mean GDP Growth 2≤INFL<5 3 2.4 7.6

5≤INFL<10 12 6.8 7.2 10≤INFL<20 5 15.5 2.1 20≤INFL<30 1 25.4 5.0

INFL≥30 12 110.6 -3.6


High inflation rate above 30 percent per annum is detrimental to growth as is associated with a negative growth of

output. For Uganda, the correlation coefficient between growth and output is negative at 0.51.On the basis of historical

statistics, for the three founding EAC member countries, there is evidence for inflation-output tradeoff. On the basis of

the correlation coefficients, the influence of inflation on output seems to be high in Tanzania and Kenya than in Uganda.

That is, growth in Tanzania and Kenya is more susceptible to high inflation, while Uganda‟s growth can withstand

relatively higher inflation. Thus, the EAC‟s across-the-board convergence criteria for inflation (5 percent) and growth (7

percent) may not be simultaneously achieved by the three member states because of the differences of inflation-growth

dynamics and economic bases.

3. Literature Review

3.1 Theoretical Literature Review

The theoretical underpinnings of inflation-growth dynamics have roots in the models of economic growth in which

Classical economics basing on supply-side theories, emphasize the need for incentives to save and invest if the nation's

economy is to grow. Keynesian theory provided the AD-AS framework which is a more comprehensive model for

linking inflation to growth. Monetarism re-emphasized the critical role of monetary growth in determining inflation,

while Neoclassical and Endogenous Growth theories sought to account for the effects of inflation on growth through its

impact on investment and capital accumulation (Brian & Howard, 2005). A detailed account of each of the growth

models in relation to inflation – growth relationship is provided hereunder.

Classical theorists laid a foundation for a number of growth theories. Classical economics hinges on the supply-side

theories, which emphasize the need for incentives to save and invest if a nation's economy is to grow; linking it to three

factors of production i.e. land, capital and labour. Classical theorists argued that growth is self-reinforcing as it

exhibited increasing return to scale. Savings were viewed as a source of investment and hence growth. Income

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

20.0

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Percen

t

Year

INFL GDP Growth


132

distribution was considered to be the most important determinant of how fast (or slow) a nation would grow; and

decline in profits were associated with competition of capitalists for workers which bid up prices rather than declining

marginal productivity of labour (Kaldor, 1956; Samuelson, 1959). Although the link between the change in price level

and its effects on profit level and output were not explicitly articulated in the classical growth theories; the relationship

between the two variables is implicitly suggested to be negative as indicated by the reduction in firm‟s profit levels

through higher wage costs.

The Keynesian models provided a more comprehensive model for linking inflation to growth under the AD-AS

framework. In the AD-AS framework, the AS curve is upward sloping in the short run so that the change in the demand

side of the economy affects both price and output (Dornbusch, 1996). Dornbusch (1996) also argues that AD and AS

yields an adjustment path which shows an initial positive relationship between inflation and economic growth but

eventually turns negative towards the latter part of the adjustment path. The initial positive relationship between

inflation and economic growth is due to the time inconsistency problem - producers feel that only the prices of their

products have increased while the other producers are operating at the same price level. The relationship between

inflation and growth is thus positive as the time inconsistency problem lures the producers into more output. Moreover,

Blanchard and Kiyotaki (1987) argue that inflation and economic growth are positively related because of the

agreement of firms to supply on agreed price with implication that the firm has to produce even at increased price.

Monetarism updated the Quantity Theory, reemphasizing the critical role of monetary growth in determining inflation.

In this framework and in the short run, money has a dominant influence on real variables (real GDP, employment and

price level); but in the long-run the influence is on price level and other nominal variables. The Philips Curve

(inflation-unemployment trade-off) holds in the short run and money is neutral and super-neutral in the long run (Tobin,

1965). Thus Monetarists suggest that in the long-run, prices are mainly affected by the growth rate in the money, while

having no real effect on growth.

Neo-classical models introduced technological change to replace investment (or capital) as a primary factor to explain

long-term growth (Solow, 1965 and Swan, 1956). Mundell (1963) articulated into the neo-classical a mechanism

relating inflation and output growth, which was distinct from the excess demand for commodities in which an increase

in inflation or inflation expectations reduced people‟s wealth. To accumulate the desired wealth, people save more by

switching to assets, increasing their price, thus driving down the real interest rate. Greater savings means greater capital

accumulation and thus faster output growth. Tobin (1965) developed the Mundell‟s theory further by introducing money

as a store of value that serves as a financial capital asset. Inflation causes productive economic agents to acquire more

capital than holding idle cash balances which leads to greater capital intensity that promotes economic growth.

Sidrauski (1967) proposed the next major development in the neo-classical models, with his seminal work on the

context of an infinitely-lived representative agent model where money is „super-neutral‟. Superneutrality holds when

real variables, including the growth rate of output, are independent of the growth rate in the money supply in the

long-run. The main result in the Sidrauski‟s economy is that an increase in the inflation rate does not affect the steady

state capital stock. As such, neither output nor economic growth is affected.

Stockman (1981) developed a neo-classical model in which an increase in the inflation rate results in a lower steady

state level of output and people‟s welfare declines. In Stockman‟s model, money is a compliment to capital, accounting

for a negative relationship between the steady-state level of output and the inflation rate. Stockman‟s insight is

prompted by the fact that firms put up some cash in financing their investment projects. Sometimes the cash is directly

part of the financing package, whereas other times, banks require compensating balances. Stockman models this cash

investment as a cash-in-advance restriction on both consumption and capital purchases. Since inflation erodes the

purchasing power of money balances, people reduce their purchases of both cash goods and capital when the inflation

rate rises. Correspondingly, the steady-state level of output falls in response to an increase in the inflation rate.

The neo-classical theoretical review demonstrates that the framework can yield very different results with regard to

inflation and growth. An increase in inflation can result in higher output (Tobin Effect) or lower output (Stockman

Effect) or no change in output (Sidrauski, 1967).

Neo-Keynesian‟s major development was the introduction of the concept of “potential output” which is the level of

output where the economy is at its optimal level of production, given the institutional and natural constraints. This level

of output also corresponds to the natural rate of unemployment, or what is also referred to as the non-accelerating

inflation rate of unemployment (NAIRU). NAIRU is the unemployment rate at which the inflation rate is neither rising

nor falling. In this particular framework, the „built-in inflation rate‟ is determined endogenously, that is by the normal

workings of the economy. According to this theory, inflation depends on the level of actual output (GDP) and the

natural rate of employment (Gordon, 1997; Roberts, 1995).

Endogenous growth theories sought to account for the effects of inflation on growth through its impact on investment


133

and capital accumulation. In endogenous growth theory, the growth rate has depended on one variable - the rate of

return on capital (Gillman, Harris & Matyas, 2002). Variables, like inflation, that decrease the rate of return, which in

turn reduces capital accumulation and decreases the growth rate. Some versions of the endogenous growth set within a

monetary exchange framework also reported that inflation rate (tax) lowers both the return on all capital and the growth

rate (Lucas, 1980; Lucas & Stokey, 1987; McCallum & Goodfriend, 1987)

On the basis of the above brief theoretical review on the inflation -growth relationship, we can discern four major

predictions on the relationship between inflation and growth as highlighted in the literature by Drukker et al. (2005).

First, some theories find that there are no effects of inflation on economic growth. Related to this category, are those

who perceive money as being super neutral (Sidrauski, 1967). Second, are those who subscribes to the fact that money

is a substitute for capital, so sees inflation as having positive effects on growth (Tobin 1965).

Third, Stockman (1981) proposes a model in which money is seen as a complement to capital, thus inflation generates

negative effects on economic growth and lastly, is a new class of theory that supports that though inflation impacts

negatively on economic growth but only when it is above a certain threshold. In these models, high inflation rates

exacerbate the frictions on financial markets, thus hampering efficiency and causing reduction on economic growth.

This class of models assumes that there is a non-linear relationship between inflation and economic growth.

3.2 Transmission mechanism from Inflation to Growth

The process through which changes in the inflation get transmitted to real GDP growth or vice versa is what is referred

to as “inflation – growth transmission mechanism”. The uncertainty associated with high and volatile unanticipated

inflation has been found to be one of the main determinants of the rate of return on capital and investment (Bruno, 1993;

Pindiky & Solimano 1993). However, fully anticipated inflation may reduce the rate of return of capital and it

undermines the confidence of domestic and foreign investors about the future course of monetary policy. Inflation also

affects the accumulation of other determinants of growth such as human capital or investment in research and

development – this channel of influence is known as the accumulation or investment effect of inflation on growth.

Inflation also worsens the long-run macroeconomic performance of market economies by reducing the total factor

productivity – this channel is known as the efficiency channel. Although in the literature it is documented that the

efficiency channel is harder to formalize in a theoretical model (Briault, 1995), its importance in the transmission

mechanism from inflation to lower growth cannot be undermined. A high level of inflation induces frequent changes in

price that may be costly to firms (menu costs) and reduces the optimal level of cash holdings by consumers (shoe

leather costs). It also generates larger forecasting errors by distorting the information content of prices, encouraging

economic agents to spend more time and resources in gathering information and protecting themselves against the

damage that may be caused by price instability – thus endangering the efficient allocation of resources.

The capital accumulation or investment effects and efficiency transmission channels from inflation to economic growth

are shown schematically in (Figure 3.1).

Figure 3.1 Transmission Mechanism from Inflation to Growth

Source: Adapted from Li Min (2006), “Inflation and Economic Growth: Threshold Effects and Transmission

Mechanisms

3.3 Theoretical Framework

3.3.1 General Growth Model

In the review of theoretical link between inflation and growth in section 3.1 we have shown the role played by growth

models in linking inflation and growth. In line with the work of Barro (1991), Levine and Renelt (1992) and


134

Sala-i-Martin (1997) the relationship between inflation and economic growth can be analyzed using the growth

regression model of the general form:

ttt XcY log (1)

Where tYlog is a growth rate in real GDP; tX is the vector of explanatory variables; is the matrix of parameters;

c is a constant and t is the error term.

There however exists a challenge of employing empirical analysis on models based on endogenous, neoclassical and

neo-Keynesian growth theories. The problem with these models is that they do not produce an exact list of explanatory

variables. For instance, the theories agree that the level of technology is an important determinant of growth, but there is

no single way to measure the technological variable. Sala-i-Martin (1997) listed such potential candidates on the role of

“level of technology” as market distortions, distortionary taxes, maintenance of property rights and degree of monopoly.

The same is true for growth determinants such as “human capital” or “efficient government”.

The choice of explanatory variables can be based on the macroeconomic theoretical framework (neo-classical growth

models would require such variables as investment, openness of the economy and population growth). The second way

to choose explanatory variables is on the basis of empirical growth literature. Levine and Renelt (1992) and

Sala-i-Martin (1997) argued that despite the existence of a large set of explanatory variables that can be potentially used

in the growth regression, only a few of them may be significant. Moreover, some variables may be significant with one

set of explanatory variables, but become insignificant with others.

3.3.2 Non-linear Model

Taking into account that inflation and growth are non-linearly related, developments in econometric modeling have

enabled authors to make a further step of not only showing the existing positive or negative relationship between

inflation and growth but also estimating the threshold level of inflation beyond, which it has a detrimental effect to

growth. The main argument here is the existence of non-linearity hypothesis, which suggest that the adverse effect of

inflation on economic growth is not universal; it appears only when inflation exceeds some turning-point or threshold

level below which inflation has a positive or non-significant impact on economic growth. In this vein, Sarel (1996),

Khan and Senhadji (2001) developed a non-linear threshold endogenous model in estimating the turning point rate of

inflation after which the inflation is deterrent to economic growth. The non-linear model takes quadratic form of:

ttt XcY 2

21l o g (2)

Where and 2 are linear and nonlinear terms of inflation respectively.

In equation (2) non-linearity is introduced in the model by including squared term of inflation,2 as an explanatory

variable. Other regressors, Xs, are added in the model as moderating variables. This is a straightforward and mostly

used technique for estimating non-linear relationships, through allowing for changes in slopes as a function of changes

in the independent variable [Pattillo, Poirson & Ricci (2002); Clements, Bhattacharya & Nguyen (2005); Devarajan,

Swaroop & Zou (1996); Hermaes & Lensink (2001)]. In this case, the slope of the estimating equation can vary with

changes in the inflation rate. This enables to observe turning points in the relationship between inflation-growth and

inflation-equality.

3.4 Empirical Evidence

The findings from empirical evidences on the inflation-growth relationship differ across countries depending on data

periods, country experiences and methodology. Among the first authors to analyse the inflation-growth relationship was

Bhatia (1960) who applied a linear model for the United Kingdom, Germany, Sweden, Canada, and Japan using

different data sets from 1812 to 1912. His findings indicate that the rate of growth was inversely related to the inflation

in Germany and Japan. In the United Kingdom, Sweden and Canada, higher rates of growth were accompanied by

higher rates of inflation. Also Dorrance (1963, 1966) and Johanson (1967) conducted a study in both developed and

developing states, respectively, and found no conclusive empirical evidence for either a positive or a negative

association between the two variables. The popular view in the 1960s was that the effect of inflation on growth was not

principally very important in both developed and developing countries.

However, change in view came after many countries experienced severe crisis of high and persistent inflation rates

associated with a general decline in macroeconomic performance and balance of payment crisis in 1970s and 1980s.

Therefore, at this time the impact of inflation to growth was given much attention and several studies were devoted to

find this relationship.

Kormendi and Meguire (1985) did a panel data study, using 47 sample countries for the period 1950-1977. They found

that an increase of inflation by 1% reduces the economic growth by 0.57 %. Fisher (1993) found negative associations


135

between inflation and growth in a pooled cross-section, time series regressions for a large set of countries. He argued

that inflation impedes the efficient allocation of resources by obscuring the signaling role of relative price changes, the

most important guide to efficient economic decision-making. Barro (1995) examined the five-year average data of 100

countries over the period of 1960-90 by using the Instrumental Variable (IV) estimation method. Using different

instrumental variables, he obtained a robust estimation result showing that an increase in average inflation by 10

percentage points per year would slow the growth rate of the real per capita GDP by 0.2-0.3 percentage points per year.

Many authors including Levine and Renelt (1992), Bullard and Keating (1995), and Bruno and Easterly (1995)

provided further facts that the negative relationship between inflation and economic growth emerges only when rates of

inflation exceed some threshold. Clark (1997) also questioned whether a uniformly negative relationship exists between

inflation and growth regardless the prevailing rate of inflation. Therefore, in the mid-1990s different studies went even

far by starting estimating not only the existing relationship between inflation and growth but also identifying the

breakpoints (threshold level) after which inflation is harmful to economic growth. They used the new class of models

regarding inflation-economic growth linkage by indicating that the relationship between them is non-linear and,

therefore, there is a threshold level. However, the empirical evidence differs substantially across the countries in terms

of the optimal level after which the inflation is deterrent to economic growth. Below are summary of some empirical

facts on the threshold level of inflation for economic growth.

Sarel (1996) used a non-linear model on a panel dataset for 87 countries covering a period of 1970-1990 with variables

on population, GDP, consumer price indices, terms of trade, real exchange rates, government expenditures and

investment rates. The empirical findings provide evidence of the existence of a structural break that was significant. The

break was estimated to occur when the inflation rate is 8%.

Christoffersen and Doyle (1998) investigated the nonlinear relationship between inflation and growth for 22 transitional

countries of Central and Eastern Europe as well as of the Post Soviet Union Countries, including Azerbaijan over the

period from 1990 to 1997. The authors found that threshold level of 13%.

Nell (2000) examined the issue whether inflation is always harmful to growth or not. The author applied the South

African economy‟s data for the period 1960-1999 and divided it into four periods. Using Vector Auto Regressive (VAR)

technique, the empirical results suggested that inflation within the single-digit zone may be beneficial to growth, while

inflation in the double digit zone appeared to impose costs in terms of slower growth.

Khan and Senhadji (2001) examined threshold effects of inflation on growth separately for industrial and developing

countries. The data set covered 140 countries from both groups and non-linear least squares (NLLS) and conditional

least squares methods were used. The empirical results verified the existence of a threshold beyond which inflation

exerts a negative effect on growth. The study established statistically significant thresholds at 1-3 percent and 11-12

percent inflation levels for industrialized and developing countries, respectively. The view of low inflation for

sustainable growth was strongly supported by this study.

Gillman, Harris and Matyas (2002) presented an econometric model with the feature of the inflation rate reducing the

return to capital, by taking two samples of OECD and APEC member countries over the years 1961-1997. Inflation rate

was included as central variable and the theory is related with the concept of equilibrium along the balanced growth

path that implicitly includes transitional approaches to the balanced growth rate. The results was consistent with Khan

and Senhadji (2000) and they showed that the effect is negative and significant at low inflation rates for the OECD.

When inflation rate going from 0-10 percent range to a 0-5 percent range, the negative co-efficient nearly doubles in

magnitude and remains highly significant.

Sweidan (2004) examined the relationship between inflation and economic growth for economy of Jordan using a

non-linear model and found a structural break point at 2 percent level of inflation. Another issue which was covered by

the study was to check the effect of inflation uncertainty on the growth and developments in the economy. The result

implied that the effects of inflation on growth were stronger as compared to the effects of inflation uncertainty and

variability. Gokal and Hanif (2004) reviewed several different economic theories to develop consensus on the inflation

and growth relationship for the economy of Fiji. Their results showed that a weak negative correlation exists between

inflation and growth, while the change in output gap bears significant bearing. The causality between the two variables

ran one-way from GDP growth to inflation.

Hussain (2005) found no definite threshold level of inflation for Pakistan but suggests that a 4 to 6 percent range of

inflation is tolerable for economy of Pakistan. This study reported similar results as the one by Singh (2003) which

recommended 4 to 7 percent range of inflation for India. Hussain (2005) followed the methodology used by Khan and

Senhadji (2001) and Singh (2003) and used the empirical results to advise the monetary authorities to keep the inflation

low and stable irrespective of any threshold level.

Ahmed and Mortaza (2005) explored the relationship between real GDP and CPI using a non-linear approach and


136

established a threshold at 6 percent level of inflation for the economy of Bangladesh. The empirical evidence

demonstrates that there exists a statistically significant long-run negative relationship between these two variables.

Pollin and Zhu (2005) presented panel regression estimates from a non-linear model form of quadratic function of the

relationship between inflation and economic growth for 80 countries over the period 1961–2000. They found threshold

inflation ranges of 14-16% for middle-income countries and 15-23% for low-income countries.

Munir et al. (2009) analyzed the non-linear relationship between inflation level and economic growth rate for the period

1970-2005 in the economy of Malaysia. Using annual data and applying new endogenous threshold autoregressive

(TAR) models proposed by Hansen (2000), they found an inflation threshold value existing for Malaysia and verify the

view that the relationship between inflation rate and economic growth is nonlinear. The estimated threshold regression

model suggested 3.89 percent as the structural break point of inflation above which inflation significantly hurts growth

rate of real GDP. In addition, below the threshold level, there is statistical significant positive relationship between

inflation rate and growth.

Frimpong and Oteng-Abayie (2010) analyzed the threshold effect of inflation on economic growth in Ghana for the

period of 1960-2008 by using threshold regression models. The result indicated inflation threshold level of 11% at

which inflation starts to significantly hurt economic growth in Ghana. Below the 11% level, inflation was likely to have

a mild effect on economic activities, while above this threshold level, inflation would adversely affect economic

growth.

Sergii (2009) investigated the growth-inflation interaction for CIS countries, including Azerbaijan for the period of

2001-2008. He found that this relation was strictly concave with some threshold level of inflation. Inflation threshold

level was estimated using a non-linear least squares technique, and inference was made applying a bootstrap approach.

The main findings were that when inflation level is higher than 8 % economic growth is slowed down, otherwise, it is

promoted.

Espinoza et al. (2010) examined threshold effect of inflation on GDP growth using a panel data of 165 countries

including oil exporting countries. A smooth transition model used over the period of 1960–2007 indicates that for all

country groups threshold level of inflation for GDP growth is about 10 percent (except for advanced countries where

threshold is much lower). They also separated non-oil exporting countries and found that inflation from higher than 13

percent decreases real non-oil GDP by 2.7 percent per year.

Also Rutayisire (2013) estimated a threshold level of inflation in Rwanda using a quadratic regression model with time

series data from 1968 to 2010. His findings indicate that the economy of Rwanda can be supported by the rate of

inflation which does not exceed 14.97 percent.

Kremer, Bick and Nautz (2013) investigated the presence of threshold effects of inflation on long term economic

growth using data of a panel for 124 industrial and developing countries. Their empirical results showed that the

estimated inflation threshold level was about 2.5% for industrial countries and 17 percent for developing countries;

above these critical levels, inflation rate leads to lower long-term economic growth rate in both cases. In addition, the

study indicated that below these thresholds, the effect of inflation on long-term economic growth was significantly

positive in developed countries; in contrast, there was no significant impact on economic growth in developing

countries when inflation is below 17 percent.

4. Research Methodology

4.1 Econometric Specification

The model used in this paper follows the Pollin and Zhu (2005) model who utilized a non-linear model form of

quadratic function to estimate the threshold level or the turning point above which inflation exerts a negative effect on

economic growth. The same approach was used by Patillo et al. (2002) and Clements et al. (2005) to estimate the

non-linear relationship between external debt and growth. Also Devarajan et al. (1996) and Hermes and Lensink (2001)

used the same methodology to determine the optimal size of government; that is the share of overall government

spending that maximizes economic growth. In line with these works, the quadratic function of equation (2) has been

adopted and modified to add more explanatory variables to examine the non-linear relationship between the rate of

inflation and economic growth.

ititititititit DUMFDIOPENFDINVPgINFLINFLYd 876543

2

210log (3)

Where d logY is growth rate of real GDP, INFL is growth rate of CPI and Pg is population growth rate, INV is

investment to GDP ratio, FD is credit to GDP ratio, OPEN is degree of openness ((Export + Import)/GDP), FDI is

foreign direct investment to GDP ratio, DUM is a dummy variable, which takes zero during price control and one

elsewhere and is the error term.


137

From equation (3), the squared term of inflation, 2

itINFL , was generated to find out the turning point, from which the

threshold level of inflation can be obtained. Therefore, combination of linear and squared term, propose that the impact

of inflation on economic growth can be described as an inverted U-shaped curve, and supports the view that the positive

effects of inflation switches to negative when inflation exceeds some threshold level. The peak of the quadratic function

identifies the inflation threshold level or the turning point above which the marginal effect of inflation becomes

negative.

The non-linearity effect of inflation on growth is assessed basing on the significance of the coefficients of linear and

non-linear estimated terms of inflation in equation (3). If both coefficients are significantly different from zero, we can

find out the peak of the quadratic function that identifies the critical point of inflation above which the marginal impact

of inflation on growth is negative. To calculate the critical point corresponding to the inflation threshold level, the

partial derivative of equation (3) is computed with respect to inflation, πt. The derivative yields the following equation

that is set equal to zero:

02

)l o g(21 it

it

INFLdINFL

Ydd (4)

Solving equation (4) for the critical point of inflation * beyond which the marginal impact of inflation on economic

growth becomes negative gives the following equation:

*INFL =

2

1

2

(5)

4.2 Choice of sample and Variables

Three founding EAC countries, Tanzania, Kenya and Uganda were used in the study for the purpose of making good

comparison. Rwanda and Burundi were not included because of very short time series (data limitation) as the two joined

the economic community in 2007. The choice of the explanatory variables in equation (3) is consistent with the choice

made in other empirical works (Khan & Senhadji, 2000, 2001; Mubarak, 2005; Risso & Carrera, 2009; Kremer et al.

2013).

4.3 Data and Data Source

The study used annual time series data set on economic growth rates and inflation from three founding EAC countries,

Tanzania, Kenya and Uganda from 1970 to 2013. The major sources of data were the Bank of Tanzania, Central Bank of

Kenya and Central Bank of Uganda and International Financial Statistics (IFS).

4.4 Unit Root Test

It has often been argued that macroeconomic data are characterized by a stochastic trend and, if unresolved, the

statistical behaviour of the estimators will be influenced by such a trend such that results may be spurious. According to

Hamilton (1994), there are different methods of overcoming the problem of spurious regression arising from using

non-stationary time series. The methods include: using a lagged endogenous variable as an explanatory variable and

differencing the non-stationary time series (until they become stationary) before variables are used in a regression. Also

another way of resolving the problem is to transform variables into growth rates, which obvious include differencing a

variable, and using ratios. For this reason, we used growth rate of real GDP, population growth rates, while the

remaining variables are ratios to GDP.

4.5 Estimation and Data Analysis Methods

An equation (3) was estimated using STATA and Gretl econometric packages using data for the three EAC member

states. Both fixed and random effect models were estimated using the Hausman specification test; and random effect vs

pooled effect models using the Breusch-Pagan test. The idea was to find out, which model is more appropriate for the

data.

In order to estimate the threshold level of inflation for each EAC member state, equation (3) was also estimated using

the Seemingly Unrelated Regression (SUR). The results for the estimated regression with the threshold values for all

three countries and individual country using the random effect and SUR model, respectively, are presented and

discussed in the following section.


138

5. Findings and Discussion of the Results

5.1 Unit Root Test

The findings from a panel unit root based on Levin–Lin–Chu test of Augmented Dicky full indicate that all variables are

stationary (Table 5.1).

Table 5.1 Levin–Lin–Chu - Panel Unit Root

t- values

Variable

Without Constant

and Trend Time Trend No-constant

dlogY -7.1408* -8.3928* -3.4274*

INFL -3.8395*** -4.5570*** -1.8675*

Pg -2.4406 -4.1370 -0.9705

INV -1.6328 -5.2774* 0.1543

FD 1.2665 0.3635 2.7699**

OPEN -3.6760 -4.6760* -0.5406

FDI 2.1777 -5.5276* -0.5935

Notes: *, **, and *** indicate rejection of the null hypothesis that panels contain unit roots at 1% , 5% and 10% levels

of significance respectively.

Source: Author‟s regression estimates

5.2 Random Effect Regression

Before estimation and discussion of the regression results of the model, we tested both fixed effect against random

effect models using the Hausman specification test, and random effect against pooled effect models using the

Breusch-Pagan test. The test in Table 5.2 indicates that the random effect model is preferred to fixed effect model,

because the P-value (p = 0.38753) is not significant for the Hausman Tests. Meanwhile since the Breusch-Pagan test is

significant (P = 0.00132) then random effect is better than the pooled model. Therefore, the random effect model is

more appropriate for the data.

Table 5.2 Fixed Vs Random Effect and Random Vs Pooled Effect Test

Breusch-Pagan test Hausman test

Null hypothesis: Variance of the unit-specific error = 0 Null hypothesis: GLS estimates are consistent

Asymptotic test statistic: Chi-square = 7.39185 Asymptotic test statistic: Chi-square = 0.74667

with p-value = 0.00132 with p-value = 0.38753

Source: Regression results

Findings from the estimated random effect model presented in Table 5.3 indicate that inflation has a positive and

significant impact on growth, while inflation squared has a negative and significant impact on growth. The coefficient

of credit to GDP ratio as a proxy for financial development was found to be positive and statistically significant.

Therefore, financially developed countries are more likely to increase growth within the EAC member states. Also

findings indicate that degree of openness has positive and significant impact to growth. This implies that trade openness

exerts leverage on the economy through increase in the market size and thus allows for increasing returns to scale. Also

it improves business competitiveness and promotes a better allocation of resources. At the same time FDI flows to

Tanzania, Kenya and Uganda have statistically significant and positive effects on growth. This suggests that most of

FDI was allocated in productive sectors, which promote growth in those countries. Dummy variable did not produce

robust results, hence was removed during estimation.


139

Table 5.3 Random Effects GLS Regression Results

xtreg dlogY lNFL INFL-1 lNFL2 NFL

2-1 Pg INV FD OPEN FDI, re robust

dlogY Coeffient Std. Error Z-statistics Prob

INFL 1.269 0.299 4.246 0.000

INFL-1 0.228 0.074 3.10 0.000

INFL2 -0.075 0.001 -3.07 0.001

NFL2-1 0.190 0.693 0.274 0.620

Pg 0.550 0.564 0.976 0.301

INV -0.027 0.048 -0.55 0.580

FD 0.083 0.027 3.03 0.002

OPEN 0.128 0.043 2.98 0.021

FDI 0.607 0.119 5.08 0.000

Cons -0.321 1.824 -0.176 0.857

sigma_u | 0

sigma_e | 2.1371770

rho | 0 (fraction of variance due to u_i)

R-square: Within =

0.851

Between = 0.920

Overall =

0.897

As it has been stated in the theoretical literature that inclusion of a large set of explanatory regressors in the

inflation-growth relationship model can be potential but only few of them may be significant (Renelt, 1992 &

Sala-i-Martin, 1997). Also according to Martin (1997) and Levine and Renelt (1992) some variables may be significant

with one set of regressors, but become insignificant with others. Findings indicate that the ratio of investment to GDP

and population are statistically insignificant.

Calculating threshold level of inflation yields:

0)075.0(2269.1)log(

it

it

INFLdINFL

Ydd (6)

From equation (6) it follows that the threshold level of inflation is:

*INFL = 8.46

)75.0(2

269.1

(7)

The above results establish that the maximum level of inflation, which supports economic growth, is 8.46%. That is, the

rate of inflation beyond the threshold of 8.46 percent for Kenya, Tanzania and Uganda has negative and significant

impact on economic growth. This is the maximum tolerable cost of inflation to growth.

The regression results of the random effect model discussed above represent the average coefficients for all three EAC

member states. The estimated threshold level of inflation of 8.46 percent assumes to support growth for all three EAC

member states. However, regression of each country is important because it provides the individual threshold level for

each member state basing on its own historical characteristics. The Seemingly Unrelated Regression (SUR) was

analysed and presented in order to give out the threshold level of inflation of each country, beyond which it has

detrimental impact on growth. The results and discussion of the SUR are presented and discussed in the next

sub-section.


140

5.3 Seemingly Unrelated Regression

Regression for the SUR treats equations for each EAC member country as independent but assuming that error terms

are related across member states. In this case, external shocks are assumed to affect all EAC member states the same

way. Therefore, there is a link among the cross-section units (EAC member states) but at the same time retaining the

coefficients for cross-section units. The results of the SUR are presented in Appendix 1 and calculation and discussion

of threshold level of inflation for each member state is presented below.

Kenya

For the case of Kenya calculation of threshold level of inflation becomes:

0)066.0(2894.0

)log( K

K

K INFLdINFL

Ydd (8)


*

KINFL = 6.77)066.0(2

894.0

(9)

Tanzania

For the case of Tanzania calculation of threshold level of inflation becomes:

0)082.0(2444.1

)log( T

T

T INFLdINFL

Ydd (10)


*

TINFL = 8.80)082.0(2

444.1

(11)

Uganda

For the case of Uganda calculation of threshold level of inflation becomes:

0)063.0(206.1

)log( U

U

U INFLdINFL

Ydd (12)


*

UINFL = 8.41

)063.0(2

06.1

(13)

Equations (9), (11) and (13) present the estimated threshold level of inflation for all three EAC member states. The

results indicate that the maximum rates of inflation which supports economic growth for Kenya, Tanzania and Uganda

are 6.77 percent, 8.80 percent and 8.41 percent, respectively, beyond which inflation starts exerting cost on growth.

6. Conclusion and Policy Recommendation

This paper estimated the threshold level of inflation, which is conducive for economic growth in the three founding

EAC countries, Kenya, Tanzania and Uganda using the non-linear threshold model for the period of 1970 to 2013.

Based on the growth model that examines inflation-growth relationship other intervening variables such as population

growth rate, investment to GDP ratio, credit to GDP ratio, degree of openness and foreign direct investment to GDP

ratio were included in the model and their impact on growth were examined.


141

Findings from the random effect model suggest that inflation has a negative and significant impact on growth, while

credit to GDP ratio, degree of openness and FDI flows to EAC member states have significant and positive impact on

growth. The average calculated coefficient of optimal level of inflation from the random effect model for all three EAC

member states suggests that the rate of inflation beyond the threshold of 8.46 percent has negative and significant

impact on economic growth. However, further findings from the Seemingly Unrelated Regression (SUR), which treats

each country separately, indicate that the optimal level of inflation for Kenya, Tanzania and Uganda are 6.77 percent,

8.80 percent and 8.41 percent, respectively, beyond which inflation starts exerting cost on economic growth. The

implication for monetary policy is that policy makers need to continue putting effort in achieving and maintaining

single level of inflation for all EAC member states to support economic growth.

References

Ahmed, S., & Mortaza, M. G. (2005). Inflation and Economic Growth in Bangladesh: 1981-2005, Working Paper

Series: WP 0604, Research Department, Bangladesh Bank, Dhaka, Bangladesh.

Barro, R. J. (1991). Economic Growth in a Cross-Section of Countries, Quarterly Journal of Economics, 106(2),

407-43.

Barro, R. J., & Sala-i-Martin (1995). Economic Growth, McGraw Hill, New York.

Barro, R. J. (1995). Inflation and Economic Growth, National Bureau of Economic Research (NBER) Working Paper

5326.

Blanchard, O. J., & Kiyotaki, N. (1987). Monopolistic Competition and the effects of aggregated Demand, American

Economic Review, 77(4), 746-66

Brian, S., & Vane, H. R. (2005). Modern Macroeconomics. Its Origins, Development and Current State. Cheltenham,

UK, Northampton, MA, USA.

Bruno, M. (1993). Inflation and growth in an integrated approach, NBER Working Paper , 4422, Cambridge,

Bruno, M. & Easterly, W. (1996). Inflation Crisis and Long-Run Growth, Journal of Monetary Economics, 41, 1.

Christoffersen, P., & Doyle, P. (1998). From Inflation to Growth - Eight Years of Transition, IMF Working Paper No.

WP/98/100

Clements, B., Bhattacharya, R., & Nguyen, T. Q. (2005). Can Debt Relief Boost Growth in Poor Countries? Economic

Issues, 34, International Monetary Fund

Devarajan, S., Swaroop, V., & Zou, H. (1996). The Composition of Public Expenditure and Economic Growth, Journal

of Monetary Economics, 37, 313-344.

Dickey, D. A. & Fuller, W. A. (1979). Distribution of the Estimators for Autoregressive Time Series with a Unit Root,

Journal of the American Statistical Association, 74, 427-431.

Dornbusch, R., Fischer, S., & Kearney, C. (1996). Macroeconomics. The Mc-Graw-Hill Companies, Inc., Sydney

Dorrance, G. S. (1966). Inflation and Growth, Staff Papers, International Monetary Fund, 13, 82-102.

Dorrance, G. S. (1963). The Effect of Inflation on Economic Development IMF Staff Papers, 10, 1- 47.

Drukker, D., Gromis, P., & Hernandez, V. P. (2005). Threshold Effects in the Relationship between Inflation and Growth:

A New Panel-Data approach, Working Paper, presented at the 11th International Conference on Panel-Data, Texas

A&M University, College Station, Texas.

Engle, R. F., & Yoo, B. S. (1991). Co-integrated Economic Time Series: An Overview with New Results, in R. F. Engle

and C. W. J. Granger, eds., Long-Run Economic Relationships, Oxford: Oxford University Press, 237-266.

Engle, R. F., & Granger, C. W. J. (1987). Co-integration and Error Correction: Representation, Estimation and Testing,

Econometrica, 55, 1-87.

Espinoza, R., Hyginus, L., & Ananthakrishnan, P. (2010). Estimating the Inflation-Growth Nexus-A Smooth Transition

Model IMF Working Paper.

Faria, J. R., & Carneiro, F. G. (2001). Does High Inflation Affect Growth in the Long and Short-run?, Journal of

Applied Economics, IV(1), 89-105.

Fischer, S., & Modigliani, F. (1978). Towards and Understanding of the Real Effects and Costs of Inflation,

Weltwirtschaftliches Archiv, 114, 810-833.

Frimpong, J. M., & Oteng-Abayie, E. F. (2010). When is Inflation Harmful? Estimating the Threshold Effect for Ghana

American Journal of Economics and Business Administration, 2(3), 225-232.


142

Gillman, M., Harris, M., & Mátyás, L. (2002). "Inflation and Growth: Some Theory and Evidence," 10th International

Conference on Panel Data, Berlin, July 5-6, 2002 D5-1

Gokal, V., & Hanif, S. (2004). Relationship between Inflation and Economic Growth, Economics Department, Reserve

Bank of Fiji, Suva, Fiji, Working Paper.

Gordon, R. (1997). The Time Varying NAIRU and its Implications for Economic Policy, Journal of Economic

Perspectives, 11(1). 11-32.

Greene, W. H. (2003). Econometric Analysis. 5th ed. New Jersey: Prentice-Hall.

Gregorio, J. (1996). Inflation, Growth, and Central Banks: Theory and Evidence, World Bank Policy Research Working

Paper 1575, Policy Research Department, Macroeconomics and Growth Division.

Hansen, B. (2000). Sample Splitting and Threshold Estimation, Econometrica, 68(3)

Hendry, D. F. (1979). Productive Failure Econometric Modelling in Macroeconomics: The Transmission Demand for

Money, in Economic Modeling: Current Issues and Problems in Macroeconomic Modeling in the UK and the USA,

(ed.). London: P. Ormerod.

Hendry, D. F. (1995). Dynamic Econometrics, London: Oxford University Press, 577-629.

Hermes, N., & Lensink, R. (2001). Fiscal Policy and Private Investment in Less Developed Countries, World Institute

for Development and Economic Research (WIDER) Discussion Paper, 2001/ 32.

Hussain, M. (2005). Inflation and Growth: Estimation of Threshold Point for Pakistan, Pakistan, Business Review, 7(3).

International Monetary Fund (2012). The East African Community (EAC): Managing the Macroeconomic Convergence

Process Ahead of Monetary Union, Washington: International Monetary Fund.

International Monetary Fund (2011). World Economic Outlook. Retrieved from

http://www.indexmundi.com/kenya/Uganda/gdp_real_growth_rate.html on 10 Jan 2012.

Johanson, H. G. (1967). Is Inflation a Retarding Factor in Economic Growth? in Fiscal and Monetary Problems in

Developing States, Proceedings of the Third Rehoroth Conference, ed. By David Krivine.

Kaldor, N. (1956). Alternative Theories of Distribution", Review of Economic Studies, 23, 83–100

Kemal, M. A. (2006). Is Inflation in Pakistan a Monetary Phenomenon? The Pakistan Development Review, 45(2),

213-220.

Khan, M. S., & Senhadji, A. S. (2001). Threshold Effects in the Relationship between Inflation and Growth, IMF Staff

Papers, 48(1).

Khan, M. S., & Schimmelpfenning, A. (2006). Inflation in Pakistan: Money or Wheat? SBP- Research Bulletin, 2(1),

213-234.

Khan, M. S., & Schimmelpfenning, A. (2006). Inflation in Pakistan. The Pakistan Development Review, 45(2), 185-202.

Khan, A. H., & Qasim, M.A. (1996). Inflation in Pakistan Revisited. The Pakistan, Development Review, 35(4),

747-759.

Levine, R., & Renelt, D. (1992). A sensitivity analysis of cross-country growth regressions, American Economic Review,

82, 942-963

Li, M. (2006). Inflation and Economic Growth: Threshold Effects and Transmission Mechanisms Department of

Economics, University of Alberta, 8-14, HM Tory Building, Edmonton, Alberta, Canada, T6G 2H4

Lucas, R. E. & Stokey, N. L. (1987). Money and interest in a cash-in-advance economy, Econometrica Vol. 55,

491–513.

Lucas, R. E., Jr. (1972). Expectations and the Neutrality of Money. Journal of Economic Theory,4, 103-l 24.

Lucas, R. E., Jr. (1973). Some International Evidence on Output-Inflation Tradeoffs. American Economic Review, 63,

326-334.

Lucas, Robert E., Jr. (1980). Two illustrations of the quantity theory of money, American Economic Review, 70,

1005–1014.

Malla, S. (2005) Inflation and Economic Growth: Evidence from a Growth Equation, mimeo, Department of Economics,

University of Hawai‟I at Monoa, Honolulu (1997). Available at www.hawaii.edu/~malla, retrieved 12 Jan 2012.

Mallik, G. & Chowdhury, A. (2001) Inflation and Economic Growth: Evidence from South Asian Countries, Asian

Pacific Development Journal, 8(1), 123-135.

http://www.indexmundi.com/kenya/Uganda/gdp_real_growth_rate.html

http://www.hawaii.edu/~malla


143

Mbelle A. V. Y, Mjema G. D., & Kilindo A. A. L (2002). The Nyerere Legacy and Economic Policy Making in

Tanzania, Dar es Salaam University Press.

McCallum, B., & Goodfriend M. S. (1987). Money: Theoretical Analysis of the Demand for Money, NBER Working

Papers, 21-57, National Bureau of Economic Research.

Mubarik, Y. A. (2005) Inflation and Growth: An Estimate of the Threshold Level of Inflation in Pakistan, State Bank of

Pakistan – Research Bulletin, 1(1-2), 35-44.

Mundell, R. (1965) Growth, Stability and Inflationary Finance, Journal of Political Economy, 73, 97-109.

Munir, Q., & Mansur, K. (2009). Non-Linearity between Inflation Rate and GDP Growth in Malaysia Economics

Bulletin, 29(3).

Munir, Q. Mansur, K., & Furuoka, F. (2009). Inflation and Economic Growth in Malaysia: A Threshold Regression

Approach, ASEAN Economic Bulletin, 26(2), 180-193.

Muwonge, A., & Obwona, M. (2003). Inflation in Uganda: Lessons from two decades, African Review of Money

Finance and Banking, JStor http://www.jstor.org/stable/23026313 .Accessed: 21/01/2014 01:25

Nell, K. S. (2000). Is Low Inflation a Precondition for Faster Growth? The Case of South Africa. Working Paper, 7.

Ng, S., & Perron, P. (1995). Unit root tests in ARMA models with data dependent methods for the selection of the

truncation lag, Journal of American Statistical Association, 90, 268-281.

Pattillo, C., Poirson, H., & Ricci, L. (2002). External Debt and Growth, IMF Working Paper, No. 02/69.

Perron, P. (1989). The great crash, the oil price shock and the unit root hypothesis, Econometrica, 57, 1361-1401.

Perron, P. (1997). Further evidence on breaking trend functions in macroeconomic variables, Journal of Econometrics,

80, 355-385.

Perron, P., & Vogelsang, T. J. (1992). Nonstationarity and Level Shifts with an Application to Purchasing Power Parity,

Journal of Business and Economic Statistics, 10, 301–320.

Pindicky, R., & Solimano, A. (1993); Economic instability and aggregate investment, NBER Working Paper no. 4380,

Cambridge.

Pollin, R. & Zhu, A. (2005). Inflation and economic growth: A cross-country non-linear Analysis.Working Paper Series

No109, PERI, University of Massachusetts Amherst.

Roberts, J. (1995). New Keynesian Economics and the Phillips Curve, Journal of Money, Credit and Banking, 27,

975-984.

Rutayisire, M. J. (2013). Threshold effects in the relationship between inflation and economic growth: evidence from

Rwanda. A paper submitted to African Economic Research Consortium (AERC). Retrieved on Monday 27 from

https://editorialexpress.com/cgi bin/.../download.cgi.

SADC (2011). Financial Systems – Tanzania, Committee of Central Bank Governors

Sala-i-Martin, X. X. (1997). I just ran two million regressions American Economic Review, 87, 178-183

Samuelson, P. A. (1959). "A Modern Treatment of the Ricardian Economy", Quarterly Journal of Economics, 73.

Sarel, M. (1995). Nonlinear Effects of Inflation on Economic Growth, IMF Working Paper WP/95/56, Washington.

Sarel, M. (1996). Non-linear Effects of Inflation on Economic Growth, IMF Staff Papers, 43(1). 199-215.

Sergii, P. (2009). Inflation and economic growth: The non-linear relationship. Evidence from CIS countries Kyiv School

of Economics

Singh, K., & Kaliappa, K. (2003). The Inflation-Growth Nexus in India: An Empirical Analysis, Journal of Policy

Modeling, 25, 377-396.

Sweidan, O. D. (2004). Does Inflation Harm Economic Growth in Jordan? An Econometric Analysis for the Period

1970-2000, International Journal of Applied Econometrics and Quantitative Studies, 1(2), 41-66.

Tobin, J. (1965) Money and Economic Growth, Econometrica, 33, 671-684.

White, H. A. (1980). Heteroscedasticity Consistent Variance Matrix Estimator and a Direct Test of Heteroscedasticity.

Econometrica, 48, 817-818.

Kremer, S., Bick A., & Nautz, D. (2013). Inflation and growth: new evidence from a dynamic panel threshold analysis,

Empirical Economics, 44(2), 861-878

Bank of Tanzania (2011). Tanzania Mainland’s 50 years of Independence. A Review of the Role and Functions of the

https://editorialexpress.com/cgi


144

Bank of Tanzania (1961 – 2011).

Appendix 1: Seemingly Unrelated Regression Results with Dependent Variable dlogY

No Country Cons INFL INFL-1 INFL2 INFL

2-1 Pg INV FD OPEN FDI R

2

1 Kenya

-0.472

(-0.15)

0.894

(3.17*)

0.894

(1.16)

-0.066

(-4.06*)

-0.031

(3.01*)

0.634

(0.50)

0.566

(4.23*)

-0.453

(-0.51)

0.205

(5.24*)

0.242

(0.64)

0.78

2 Tanzania

1.522

(0.77)

1.444

(4.05*)

0.282

(-3.00*)

-0.082

(-2.84**)

-0.040

(-3.52*)

-4.267

(-1.99**)

0.044

(0.27)

0.114

(3.64*)

0.738

(2.10**)

0.502

(4.69*)

0.90

3 Uganda

1.320

(0.19)

1.06

(2.88**)

0.966

(-1.98**)

-0.063

(-3.65*)

-0.018

(-3.40*)

-0.299

(-1.27)

0.911

(1.39)

0.449

(2.12**)

0.061

(2.38**)

0.927

(2.10**)

0.81

Note: * and ** means statistically significant at 1% and 5%, respectively.

Source: Authors findings.

This work is licensed under a Creative Commons Attribution 3.0 License.

http://creativecommons.org/licenses/by/3.0/

http://creativecommons.org/licenses/by/3.0/

Appropriate Threshold Level of Inflation for Economic ...

Documents

Transcript of Appropriate Threshold Level of Inflation for Economic ...